Monash University researchers are challenging one of modern medicine’s most persistent assumptions: that shoulder pain is usually caused by a structural defect that must be identified and repaired. In a review published in JAMA Internal Medicine, the researchers conclude that most people with shoulder pain can recover through education, activity modification, symptom relief and time, without early scans or invasive treatment. The findings could reshape how one of the most common musculoskeletal complaints is managed in primary care, where patients are frequently referred for X-rays, ultrasounds or magnetic resonance imaging before the underlying problem has been clearly understood.
Shoulder pain affects almost one in six adults each year and is the third most common musculoskeletal reason for visiting primary care, after back and knee problems. Although the shoulder is a complex joint involving bones, muscles, tendons, ligaments, bursae and nerves, pain in the region does not always correspond to visible damage in one of those structures. The review argues that this mismatch is central to the problem. Imaging can reveal tendon thickening, small tears, bursal changes or bone irregularities in people who have no pain at all, particularly as they age. Finding one of these changes may therefore create the impression of a specific injury even when it is simply part of normal biological variation.
The researchers recommend that the first clinical assessment should focus on the patient’s symptoms, medical history and physical examination rather than immediately searching for an anatomical explanation. Clinicians should first rule out serious conditions, including infection, cancer and fractures, while also checking whether the pain is actually coming from somewhere else. Nerve irritation in the neck, for example, can produce pain around the shoulder, while some heart or chest conditions may cause discomfort that is felt in the shoulder or arm. Once these possibilities have been considered, many patients can be safely managed without routine imaging, particularly when there are no warning signs suggesting a serious underlying disease.
Lead author Dr Romi Haas, a Research Fellow at the Monash School of Public Health and Preventive Medicine, said the traditional model of care often treats the shoulder as if every painful movement signals a mechanical failure. “The traditional approach has been to look for a structural flaw by scanning the shoulder, assuming that if you can identify the source of the pain, you can fix it,” she said. Yet the review indicates that structural findings frequently have a weak relationship with pain intensity, disability or long-term recovery. Tendons can show age-related changes without causing symptoms, just as skin develops wrinkles or hair turns grey without representing a disease requiring correction.
This is particularly relevant to subacromial pain, the most common pattern of shoulder pain seen in primary care. It is generally felt over the upper or outer shoulder and may become worse when a person lifts the arm, reaches overhead or lies on the affected side. In the past, symptoms of this kind were often assigned labels such as bursitis, impingement or rotator-cuff tendon damage. Although such terms may appear precise, they can suggest that tissues are being trapped, inflamed or torn in a way that must be physically repaired. The review supports using more neutral, location-based language such as “subacromial pain,” which may reduce fear and help patients remain active while symptoms settle.
The distinction matters because pain is not a direct readout of tissue damage. It is produced by the nervous system through the interaction of signals from the body, previous experiences, expectations, stress and the brain’s assessment of threat. A scan can identify an abnormality, but it cannot by itself establish that the abnormality is the cause of pain. When people are told that a tendon is torn or that bone is compressing a tendon, they may understandably protect the shoulder, avoid movement and become concerned that ordinary activity will make the condition worse. That response can contribute to weakness, stiffness and persistent disability, even when gradual movement would be safe and beneficial.
The review also questions the value of a commonly performed shoulder operation that removes bone and soft tissue to create more space around the tendons. High-quality trials have found that this procedure offers no meaningful long-term advantage over a placebo operation in which patients undergo skin incisions but the intended tissue removal is not performed. Co-author Dr Thomas Ibounig, a shoulder and elbow surgeon at Helsinki University Hospital, said the evidence shows that pain is not always caused by an anatomical problem that surgery can simply correct. The conclusion does not mean that every shoulder operation is inappropriate; surgery may still be considered for selected injuries or serious conditions. It does mean that routine procedures for uncomplicated shoulder pain require careful justification.
Corticosteroid injections occupy a similar position in the evidence. They may provide short-term pain relief, which can sometimes help a person begin rehabilitation or sleep more comfortably, but their benefits are generally limited over time. The review says injections do not routinely need to be guided by ultrasound or another imaging technique. When the injection is administered by a clinician who uses the patient’s anatomy and physical landmarks, outcomes are comparable in many cases, while the approach avoids the extra cost, scheduling burden and technical demands associated with image guidance. The finding reflects a broader principle in evidence-based care: a more elaborate intervention is not necessarily more accurate or more effective.
For patients, the recommended approach may feel less dramatic than receiving a scan or a procedure, but it is designed to be active rather than dismissive. Treatment can include an explanation of the likely course of symptoms, advice to continue ordinary activities within tolerable limits, temporary modification of painful movements, exercises to restore strength and mobility, and medication such as anti-inflammatory drugs when appropriate. Some people may improve with watchful waiting and reassurance alone. Professor Rachelle Buchbinder, a rheumatologist and clinical epidemiologist at Monash University and the review’s senior author, said most people recover within a year and often much sooner. She cited long-term research involving 526 adults with new shoulder concerns, in which almost 80 per cent reported satisfactory improvement through approaches including watchful waiting or anti-inflammatory medication.
The researchers say the review is intended as a practical guide for general practitioners, physiotherapists and surgeons, rather than a rigid rule for every patient. Persistent or worsening symptoms, major loss of function, significant trauma or signs of systemic illness may require additional investigations or specialist care. For the majority of uncomplicated cases, however, the evidence points away from a scan-first model and toward communication, movement and time. The message is potentially far-reaching: when clinicians explain that many scan findings are normal age-related changes and that pain does not automatically mean damage, patients may avoid unnecessary anxiety, procedures and costs while receiving care more closely aligned with how recovery actually occurs.
Subject of Research: People
Article Title: Management of Shoulder Pain in Primary Care: A Review
News Publication Date: 17-Aug-2026
Web References: https://doi.org/10.1001/jamainternmed.2026.3135; https://research.monash.edu/en/persons/romi-zaslavsky/; https://www.monash.edu/medicine/sphpm/home
References: JAMA Internal Medicine, DOI: 10.1001/jamainternmed.2026.3135
Keywords: shoulder pain, musculoskeletal health, medical imaging, self-management, physiotherapy, primary care, corticosteroid injections, shoulder surgery, subacromial pain, evidence-based medicine
